Why should you consider using AI tools as part of your SEO strategy?
AI is rapidly becoming a standard tool in SEO strategy – it’s not just about keeping up with the latest technology, it’s about keeping pace with competitors in your space—AI is a powerful tool for supporting your work and letting you focus on activities that should have a human element. Generative AI (focused on large language models), can automate and streamline many aspects of SEO, freeing you and your team up to focus on other important tasks.
AI can provide a perspective you may not have considered – it can help you avoid tunnel vision and continually reassess your assumptions around your SEO strategy. If applied properly to your traffic data, it can help you understand your audience and better tailor your SEO strategy to their needs.
How should you expect generative AI tools to impact search engines and click throughs generally?
AI has allowed for the production of vast amounts of content – AI is making creation much easier and more efficient. The advent of generative AI significantly altered the landscape of SEO.
Quality, authority, and trustworthiness are still key – while generative AI has made it easier to produce large volumes of content, the focus for SEO shifted from quantity to quality. Google and other search engines continue to prioritize content that is authoritative and trustworthy. Therefore, it’s crucial to ensure that your content is meaningful and provides value to the reader.
There’s more content than ever, so indexing is becoming more competitive – with the increase in content available, Google has more than ever to crawl. Ensuring that your best content gets indexed by search engines has become more critical. To achieve this, you need to effectively communicate to search engines why your content is important and why you are an authoritative voice on the topic.
Niche and expert content is more important – generative AI has allowed brands to focus on producing more specialized content. They no longer need to produce basic or foundational content and can instead concentrate on being experts in their specific fields.
AI may begin to move search behavior over from Google – there has been a noticeable shift in how users search for information, with younger users increasingly bypassing Google and turning to platforms like Amazon, TikTok, or Pinterest. However, trust in AI tools as a source of information is not there yet due to limitations, particularly in current events.
How should you adjust your SEO strategy to account for the changes to the search landscape?
Publish with E-E-A-T (Experience, Expertise, Authority, Trustworthiness) in mind – demonstrate your expertise and authority on the topics that matter to you and your seller. Google places a huge emphasis on verified reviews and testimonials. Therefore, involving your customers in the conversation on your website in an authentic way is crucial.
See a comprehensive guide from Lily Ray on E-E-A-T here.
Be transparent and authentic with your content – being transparent about your content (e.g., whether it’s an affiliate post or sponsored content) is beneficial. Audiences and Google respond better to transparency. Brands should be authentic in what they’re discussing.
Pruning content is an important strategy if you have more than 100,000 pages – at that point, you want to ensure you focus on what truly matters to your website and brand. Reduce the fluff and ensure Google spends time on your most important content by considering your “crawl budget”.
Craft intuitive navigation and information architecture – your website’s navigation should support what matters most on your site. Pointing to your priority content is essential to ensure Google knows what’s important to you.
Spend time on schema markup – provide Google with more information about your content through schema markup to improve your SEO. Schema can be used to mark up author information, articles, learning resources, and videos. Schema markup is a closed language set that Google can understand quickly, making it easier for them to index your content. Generative AI can even be used to create scripts for structured data, making it easier for less technical individuals to implement this technical aspect without needing to involve developers.
Let go of content that doesn’t make sense – it’s important to have an honest conversation about what you should rank for and what you shouldn’t. Let go of areas that don’t make sense for your brand to focus on. SEO won’t solve the problem of trying to rank for something that’s not aligned with your brand’s expertise or authority.
What tools should you consider using when leveraging AI in SEO?
| Category | Tools | Use cases | ||
| General LLMs | ChatGPT-4 | For content analysis and summaries – GPT-4 is an excellent tool for analyzing existing pages. It can crawl through pages and provide summaries, ensuring that the spirit of the article is retained even after it’s written. | ||
| Point Tools | Copy.ai | For on-brand content creation – Copy.ai is a great tool for creating content that aligns with your brand. You can input your brand values, tone, and voice, and the tool will generate content in your style. It’s especially useful when you want to edit content rather than create it from scratch. | ||
| Test IQ & COG-Bias | For bias identification – these tools are useful when creating surveys to gauge customer sentiment. They can identify potential biases in your language, helping you improve your prompt engineering and communication. IQ Disc is especially helpful as it identifies the types of biases present. | |||
| Claude | For document summarization – Claude is a handy tool for summarizing documents. It’s particularly useful when you need to consume a lot of information at once. | |||
| Midjourney | There are numerous tools available for creating video content – some can even emulate your voice from recordings, allowing you to create videos without spending time in a production studio or in front of the camera. | |||
| Machine Learning Tools | Cornell’s Tools | For data analysis – Cornell has published a variety of machine learning models that you can use to analyze your datasets. These models can help with sentiment analysis and topic modeling, allowing you to distill a large amount of information into what really matters—you can put in 180K keywords and use a model to break down into 18 main topics | ||
What are some common use cases for leveraging AI in SEO to generate content?
| Running your Blog / Generating Organic Traffic | |||
| What it is | Use AI as a drafting tool – AI tools like ChatGPT are best used for drafting content. However, it’s crucial to include human curation to ensure the accuracy of the information. Google’s algorithms can detect factually incorrect content, which can lead to a decline in your visibility. | ||
| Tips to get right | Be aware of violating other’s copyrights – AI tools have been trained on a vast amount of internet content, including copyrighted material. This raises potential copyright issues. For instance, if ChatGPT generates content that unknowingly infringes on a copyright, it’s unclear who is responsible – the AI provider or the brand that used the AI. Until this is resolved, it’s best to use AI tools with caution. Anything crucial to your brand should be owned by a human – key elements like a logo or a tagline, should be created or owned by a human throughout the process. This is because generative AI-created content is not protected by copyright laws. Be cautious with proprietary information – if you’re inputting proprietary information into an AI tool, be aware that once you submit it, it may no longer belong to you. This could lead to leaks of sensitive information. Use AI for non-sensitive tasks – AI tools can be useful for non-sensitive tasks, like drafting blog posts. However, for sensitive tasks that affect your business, like website migrations, it’s best to avoid using AI tools. Monitor performance vs. human-generated content – the performance of AI-generated content versus human-created content is still being evaluated. AI tools are getting better at sounding natural, but they can still make errors. Users are also becoming more educated and can often recognize AI-generated speech patterns. | ||
| Tools | Choose the right tool for your needs – the best tool for you depends on your specific needs: – If you want a tool to learn your style → Copy.ai – If you want a tool to protect proprietary data → Copy.ai – If you want a tool with a robust user community → ChatGPT – If you want a tool with a low barrier to entry → ChatGPT – If you want a tool that cites its sources → Perplexity AI | ||
| Tailoring custom landing pages for SEO and SEM | |||
| What it is | Curate for SEO, Generate for SEM – when tailoring custom landing pages, it’s recommended to curate and be selective and scrutinizing for SEO. However, for SEM, you can leverage generative AI to create thousands of different landing pages. This allows you to target a wide array of audiences with more personalized ad content. | ||
| Tips to get right | Create a template and adapt it for different industries – being able to create ad copies and headlines that speak to the individual will make it more powerful. Generative AI can bring down the time necessary to create copy or headlines that resonate with each individual business owner. Understand your audience’s concerns – use tools like Chat-GPT to understand the unique concerns of each industry you’re targeting. While all businesses aim for sustainability, each industry will have its own unique challenges and frustrations. By understanding these, you can create more meaningful and personalized content. Curation of generated content is key – despite the power of AI, curation remains a crucial part of the process. It ensures that the generated content is relevant, meaningful, and effective in reaching your target audience. | ||
| Creating Social Media Posts | |||
| What it is | Use AI to repurpose existing content into social media assets – AI can help you repurpose content such as emails, newsletters, or blog posts into social media posts. For example, you can feed a blog post into an AI tool and ask it to pull out quotable content, next steps, or action items. This can be done with various types of content, including resources and meeting notes. | ||
| Tips to get right | Create a content calendar – by repurposing your existing content into social media posts, you can effectively create a content calendar. This allows you to maintain a consistent posting schedule, while ensuring that your social media content aligns with your overall messaging and brand. Experiment with tone and style – AI tools can also help you experiment with different tones and styles for your social media posts. For example, you can ask the AI to generate variants of a post with a humorous tone or to create an April Fool’s joke. Evaluate your content while creating posts – using AI to generate social media posts can also provide valuable insights into your content. If the AI doesn’t highlight a point that you thought was important, it might indicate that you need to emphasize that point more in your original content. | ||
| Tools | AI tools like Mid Journey and Dall-E can be used to generate images – you can automatically generate imagery to accompany your social media posts or produce net new assets. This can result in visually appealing posts that you can edit as needed. | ||
Other content generation use cases to consider include:
- Repurposing existing content in new forms – AI can be a powerful tool for repurposing existing content. For instance, if you’ve created a resource article or a how-to guide, you can use AI to generate a document, social media posts, or even a script for a promotional video based on that content. This not only saves time but also ensures consistency across different content formats.
- Tailoring content to readers – AI can also be used to create variations of your existing content. For example, if you’re writing newsletters, AI can help you generate different versions to cater to different audience segments. This can make your content more engaging and personalized.
- Idea generation – AI tools like Chat GPT can be used to generate content ideas. By inputting questions related to your target audience’s concerns and frustrations, you can uncover new insights and validate your assumptions. This can help you create content that addresses your audience’s needs more effectively.
- Persona Research and Creation – AI can be a valuable tool for persona research and creation. By analyzing aggregated user data, AI can help you understand what your audience is consistently talking about. This can provide valuable insights into your audience’s needs and preferences, helping you create more targeted and relevant content.
In what ways can Generative AI enhance the understanding of search intent and user behavior for SEO professionals?
| Working with Metadata | |||||
| What it is | AI can save time on metadata creation – writing metadata and alt text can be a time-consuming task, especially for large websites. AI can generate this information in a matter of minutes, saving you a significant amount of time. However, it’s important to review the AI-generated metadata and alt text to ensure that it’s accurate and appropriate. | ||||
| Tips to get right | It can help you with your schema and alt text- Schema is a type of structured data that helps search engines understand the content on a web page. Alt text is used to describe images for people who can’t see them. Both of these are time-consuming endeavors and can be made easier with the help of AI. | ||||
Notes on alt text:
AI can generate wild descriptions – when using AI to generate alt text, it can sometimes produce very unusual or incorrect descriptions of images. Therefore, it’s important to always review and curate the AI-generated alt text before using it.
Alt text usage varies among internet users – there are many different ways that people on the internet use alt text. Some people use it for SEO purposes, while others use it for accessibility. It’s important to understand the different ways that alt text can be used and to use it in a way that best serves your needs.
Note: AI can reveal new product uses when generating descriptions – when AI is used to generate product descriptions, it can sometimes reveal new ways that people are using your products. This is because AI uses predictive text based on a large amount of data, which can include information about how people are using your products in ways that you hadn’t considered.
Notes on schema:
AI can help you create the scripts to put in your structured data – Google accepts 1-2 dozen types of structured data on your content—it’s all based on schema.org. This data can provide a detailed description of the page and its main content.
It can help put structured data for videos – Google can understand videos, but they’re not going to spend the resources to do so. You can add structured data to serve that content up to them and make it easier for them to consume. If you’re less technical, AI can create the data for you and you can get a technical implementation of it without having to bug your developers.
| Pulling insights from traffic data | |||
| What it is | Using AI for data analysis – tools like Claude, ChatGPT, or ML models can simplify the process of data analysis. You can input reports or datasets and ask the AI, “What’s interesting about this?” This can save you hours of manual data digging. However, these tools typically perform better with numeric data rather than visual data. | ||
| Tips to get right | With large datasets, consider using the free machine learning models from Cornell – with some time spent on their tutorials, you can start loading in your data, finding unique insights, and organizing your data in a comprehensible manner within a week. Their resources make it easy to extract meaning from your data. | ||
Examples of pulling insights from traffic data using AI:
- AI can be used for Metadata and Alt-Text analysis – this process can be time-consuming, but it’s crucial for understanding user behavior and search intent. However, keep in mind that Google may alter your metadata about 60% of the time.
- Persona discovery and research – understanding your audience’s intent and motivations is a key part of persona discovery and research. Machine learning models can assist in this process, although some models may be more effective than others. For instance, GPT-3 may not be the best choice for this task.
How do you think about your own content showing up in AI-generated results?
Concerns about rankings within AI systems are growing – many brands are curious about how often their content shows up in AI systems like ChatGPT. They want to know when they’re being cited or when information from their website is being used. This is particularly important for organizations that deal with sensitive information, such as government clients in the financial space. Misquotation can lead to serious consequences.
Currently, there’s no toolset available to track when your content is being used – or when your brand is being promoted within AI systems. This lack of transparency makes it difficult for brands to understand and control how their content is being utilized.
SEO strategies still apply to increasing your relevance to AI tools – to increase the chances of your content being picked up by AI systems (in the case that, for example, someone asks an LLM for five Marketing Automation tools they should evaluate), continue implementing SEO strategies. Encourage conversations about your brand, gather feedback from your customers, and present your content in an understandable way.
AI system owners will write the rules in an LLM-dominated future – just like how Google dictates the rules of SEO, the owners of AI systems like OpenAI and Microsoft are writing the rules for how these systems operate. Brands are beholden to whatever these companies decide to give them. If you’re an e-commerce brand, think about the different ways people search for your products and offer various filter possibilities. The goal is to make it easier for users to find the answers they seek without having to go through multiple steps.
How do you think about scale vs. quality of content when using generative AI in SEO?
Quality trumps quantity in SEO strategy – especially now, when crawling and indexing can no longer be taken for granted. It’s better to produce one high-quality, high-authority piece than ten mediocre ones.
Repackaging and redistributing content is key – it’s not always about generating new content. AI tools can help repackage existing content for continued distribution. The effort put into content generation should be matched, if not exceeded, by the effort put into its distribution and repackaging. Ideally, you should be able to create one high-quality piece and spend three solid months promoting it before needing to create another one.
Crawling and indexing is something you can no longer take for granted – you have to write quality content that’s squarely within your area of expertise in order to appear alongside relevant search terms.
Focus on your area of authority – trying to come up with new messages all the time can lead to venturing into areas where you’re not the authority. It’s better to focus on what you know and are an authority on. Even if it seems basic or elementary to you, it’s valuable because it’s your area of expertise.
How can you align your prompt engineering with good SEO practices and strong branding?
Prompt engineering is a tool, not the entire solution – while prompt engineering can be a powerful tool, it’s important to remember that it’s not the be-all and end-all. It’s a tool that can be used to generate text, but it’s not the only tool in your toolbox.
Prompt engineering is even more important in image searches – when using prompt engineering for images, take the time to consider the art styles, colors, and feelings you want your imagery to evoke. This can greatly enhance the effectiveness of your prompts.
Always ask for multiple variations – one of the best practices in prompt engineering is to always ask for multiple variations. This allows you to combine and piece together different elements to create the perfect prompt.
Consider your audience and tone – always keep in mind who will be reading and consuming your content. The tone and language used should be tailored to your audience. For example, the language used for a mommy blog will be different from that used in a Harvard Law Review.
Educate your AI model – just as you would brief a human writer, you should also educate your AI model. Provide it with an outline, the major points you want to make, your target audience, and how you want your audience to feel after reading the article.
As search engines become more adept at understanding natural language, what are the implications for SEO professionals in terms of optimizing for voice search?
Voice Search Optimization is not a concept new to LLMs – it’s something that many brands have been thinking about over the last five years. With the rise of at-home devices like Siri and Alexa, the use of voice search has already seen a significant increase.
To maximize your voice-search positioning, create content that’s in your audience’s idiom – Google and other search engines have been striving to understand natural language better and service queries from voice users. To improve your presence via voice search, communicate with your prospects, customers, and users in a way that resonates with them. If you’re an exterminator, don’t write at a PHD level as you’re primarily dealing with homeowners. Be true to your brand and understand how you want to present to your audiences.
ChatGPT-generated content will have voice search limitations – because ChatGPT generates content in a somewhat generic voice it may not be the best tool for voice search optimization.
What are the legal considerations of using generative AI in SEO? When do you need to disclose the use of AI?
Disclosure of AI use is context-dependent – whether or not you need to disclose the use of AI in content creation depends on the context. If the content is largely generated by the AI, it may be important to disclose this, especially when dealing with facts or reporting on events. However, in the B2B space, this is not yet a common practice.
What are the most important things to get right?
Human curation is crucial in leveraging AI and SEO – recognize that errors can occur and ensure your brand is well-represented. If a piece of content is significant enough to your brand that you would want to copyright it, it’s best to keep AI out of the process.
Understanding and leveraging keywords and naming ideas – LLMs are known for their effective use of language. However, there have been instances where products are named without proper research into what the name signifies. It’s crucial to Google and understand the meaning of a product name before finalizing it. This can prevent potential misunderstandings or misinterpretations that could negatively impact the product’s reception.
What are common pitfalls?
Skipping out on human editing – it’s important for a human to thoroughly edit your content. When there’s a lack of editing, it becomes obvious and can negatively impact your brand’s reputation. While tools can help shorten your processes, they should not replace them. It’s still important to put in the effort to create high-quality, original content.
Foregoing fact-checking – Fact-checking is a crucial part of content creation. Publishing content without verifying the information can lead to misinformation and damage your brand’s credibility.
Blatantly disregarding copyright – brands sometimes publish content that clearly rips off another brand’s work. This not only shows a lack of originality but also can lead to legal issues.
Responses